SI.com: Romo shoots down doubters with attack on passing efficiency records

[Everson24]

I really wish people would stop saying Romo played bad in the 2007 playoff game against the Giants. IT IS SIMPLY NOT TRUE!!!!!! I dare anyone to go back and watch that game and tell me Romo played a bad game. If anyone can watch that game and STILL say Romo played bad, they have ZERO football IQ. ZERO!!!!!!!

I agree 100%. The defense that gave up a TD with less than a minute to go in the half, the special teams who gave up the long punt return to set up the Giants winning score, the offensive line and backs who could not pick up a blitz and Patrick Crayton who somehow chose this game to have hands of stone get a free pass but not Tony Romo! Even with all those penalties and failings by the above mentioned group, We still would have won the game if Crayton doesn't stop his route on the next to the last play of the game.

If Crayton runs his route properly, Tony is known as a great comeback quarterback ala Ben Rothlisberger. Frustrating to say the least. I can only imagine how Romo feels about all this 0-2 playoff talk.

 

[kmd24]

The mistake with passer rating that most people make is trying to use it as a proxy for valuing QB's. IOW, Chad Pennington is better than Roger Staubach because his career passer rating is 6 points higher. That's foolish.

For one, QB rating is conflated with a team's passing attack. The same QB with better protection, better receivers, a better running game, etc. will end up with a better passer rating.

A second factor is the changing rules over the course of football history. The modern game is much more friendly to QB's and WR's than it was in the 60's and 70's. It is a mistake to compare QB's from different eras via passer rating if your goal is to differentiate the QB's by their individual ability.

If the question you want to answer is whether passing attacks have become more efficient over the years, then QB rating is a very good statistic. Only a fool would argue that the passing game was more effective in the 1970's than it is today, and that is reflected in the gradual rise of the average passer rating over the years. When the statistic was devised in 1973, it was intended to yield an average rating of 66.6. The average passer rating in 2007 was 83.5.

It's frustrating to see people arguing past each other about this statistic, and it's really funny to see such vitriol towards a number. It's just a statistic, ostensibly calculated to give some guidance about what kind of raw statistics would constitute an effective passing attack, and it's very good at that, as theogt has mentioned. It's when we misguidedly try to answer the wrong question that things go awry.

As far as the statistic being antiquated, I think that argument has validity on some level, but I question the motivation. As theogt mentioned in another thread, there has been remarkable consistency in the variance of passer statistics over the years, despite some drift in the centering. For example, completion percentage has gradually increased over the years, and interception percentage has dropped. But most often, people want to monkey with the stat to reorder quarterbacks in a way that meets their preconceived notions, without regard for the new statistic's predictive (or even descriptive) ability.

 

[AmericasTeam31]

Wow! I guess I never really thought about it, but, outside of December, the Cowboys have only lost 4 games when Romo is the starting QB!!!! That is impressive enough, but 8 of his 12 career regular season losses have come in December... I'm just saying I never realized that stat....

 

[DBOY3141]

Marino was 8-11 in playoff games and missed the playoffs 6 times in 17 years.
He is one of the greatest QB's ever, but never won the big won. Romo could be on that same track, regardless if he ever wins a playoff game, he is a top QB, because every week the Cowboys have a chance to win, mostly because of Romo.

The media are a bunch of clowns who care more about drama than facts. Poll every GM in the league, and they will rank Romo as a top 5 QB.

 

[Bob Sacamano]

Romo is truly special

 

[wileedog]

There is no way a QB who goes 12-13 and gains 80 yards should have a higher passer rating of a QB who goes 11-20 and gains 200 yards.

I'm not sure I agree with that completely.

Those 12 completions might have kept several drives alive.

Those 9 incompletions might have forced multiple punts.

Completion% is one of the first things I look at when I evaluate a QB. There are very, very few elite QBs with a low comp rate, especially in the modern age where the passing game has become the norm. Conversely, I think its been an achilles heel to a guy like McNabb who despite playing in an extremely QB friendly system has a career % <60. Its one of the reason's I consider him a rung below the top shelf guys. Eli as well.

I do think the system needs tweaking, but I don't think the QB who dinks and dunks down the field efficiently to score TDs is any worse than the guy who flings the ball all over the field and happens to connect on a bomb or two. In fact the guy who can do both is probably the best of the three, but I have no idea how you would capture that from stats in a holistic way, outside of the current system.

 

[BAT]

The mistake with passer rating that most people make is trying to use it as a proxy for valuing QB's. IOW, Chad Pennington is better than Roger Staubach because his career passer rating is 6 points higher. That's foolish.

For one, QB rating is conflated with a team's passing attack. The same QB with better protection, better receivers, a better running game, etc. will end up with a better passer rating.

A second factor is the changing rules over the course of football history. The modern game is much more friendly to QB's and WR's than it was in the 60's and 70's. It is a mistake to compare QB's from different eras via passer rating if your goal is to differentiate the QB's by their individual ability.

If the question you want to answer is whether passing attacks have become more efficient over the years, then QB rating is a very good statistic. Only a fool would argue that the passing game was more effective in the 1970's than it is today, and that is reflected in the gradual rise of the average passer rating over the years. When the statistic was devised in 1973, it was intended to yield an average rating of 66.6. The average passer rating in 2007 was 83.5.

It's frustrating to see people arguing past each other about this statistic, and it's really funny to see such vitriol towards a number. It's just a statistic, ostensibly calculated to give some guidance about what kind of raw statistics would constitute an effective passing attack, and it's very good at that, as theogt has mentioned. It's when we misguidedly try to answer the wrong question that things go awry.

As far as the statistic being antiquated, I think that argument has validity on some level, but I question the motivation. As theogt mentioned in another thread, there has been remarkable consistency in the variance of passer statistics over the years, despite some drift in the centering. For example, completion percentage has gradually increased over the years, and interception percentage has dropped. But most often, people want to monkey with the stat to reorder quarterbacks in a way that meets their preconceived notions, without regard for the new statistic's predictive (or even descriptive) ability.

But isn't this the reason the passer rating was developed in the first place? The powers that be needed a stat/calculation to compare/contrast QB production and efficiencies? Ranking is a natural extrapolation of that data. Even ranking by era, both Young and Favre are ranked above Aikman. Aikman was much more efficient in the postseason and more importantly he was the most successful QB due to the number of championships he won.

I agree and understand your point w/rule changes, but I just feel that a true "QB" rating has to take more factors into consideration. The current system is one dimensional in that it only evaluates a QB's passing numbers, further it places to much emphasis on completion over yards gained, and no weight given to yards achieved on the ground. And there must be a way that a completion for no gain (or negative gain) does not equal a completion for a 30 yard gain. And should the QB rating incorporate wins too? Or are W-L a separate but equal stat (that seems to be forgotten/unused during rankings/comparisons). So instead of the system being "predictive" it could more "conclusive".

One final note, the current passer rating system has a ceiling of 158.3. You have admitted that the average QB rating is trending upwards with the new rules and other factors that have increased passing efficiencies. As this trend increases with corresponding increases to the average QB ratings, what happens to that ceiling? Is it subjectively raised? Brady has come close to reaching it already when he had that near perfect game a while back, in the 140's if I recall.

Just b/c a tool is 50% effective does not mean it cannot, should not be improved. Even if the tool was 90% effective/efficient (which the current passer rating does not even come close to), improvement should still be strived for. Improvement is improvement, motive or my "vitriol" notwithstanding.